01:15 < bridge> i thought chillerbot-zx was the illegal one, why is it on ger10 08:15 < bridge> kinda, check out my new ddnet testmap, Mycelium 08:41 < bridge> It is the illegal one. So psst before the admins notice him .\_. 10:00 < bridge> do we have websocket support in ddnet 10:00 < bridge> maybe i can connect the client to the event feed 10:00 < bridge> pog 10:17 < bridge> @milkeeycat how is gentoo going 10:17 < bridge> yes 10:17 < bridge> @fokkonaut wasnt it removed 10:17 < bridge> true 10:17 < bridge> xd 10:17 < bridge> but adding a external websocket connection shouldnt be hard 10:17 < bridge> i dont mean a server one 10:17 < bridge> ye 10:17 < bridge> klient 10:18 < bridge> :eyes_anger: 10:18 < bridge> most events are from chinese 10:18 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1159041826771177482/image.png?ex=651e7151&is=651d1fd1&hm=12dd3ba5f2860b497b845519dbcdb320f7a4761f4284fb4b74c2a2976236be3c& 10:18 < bridge> xd 10:19 < bridge> well most players are from chinese 10:19 < bridge> china* 10:19 < bridge> u should speak english correctly ur american 10:20 < bridge> :justatest: 10:20 < bridge> :justatest: 10:20 < bridge> https://github.com/edg-l/teeobserver/blob/457c4c7bf465001ec0776255fdeffd7d002e56df/src/main.rs#L102C26-L103 10:20 < bridge> broken 10:20 < bridge> 128 10:20 < bridge> https://github.com/MilkeeyCat/ddnet-web-demo-viewer 10:20 < bridge> Now u can get data about demos and maps in js. For what? Why? Idk but you can :p 10:20 < bridge> 128? 10:20 < bridge> @ryozuki max clients in the browser is at 128, not 64 10:20 < bridge> oh 10:21 < bridge> the master itself can take infinite, i believe, but the client can take up to 128 10:21 < bridge> @fokkonaut its not broken though 10:21 < bridge> probs 10:21 < bridge> that just specifies the initial capacity 10:21 < bridge> hashset can grow 10:21 < bridge> Ah 10:21 < bridge> its so it doesnt reallocate 10:21 < bridge> when adding 10:21 < bridge> default initial is 4 or 8 iirc 10:21 < bridge> i thought that ddnet could only show 64 players? 10:21 < bridge> i think 64 at a time 10:21 < bridge> but u can hack it 10:21 < bridge> and servers are usually 64 with the exception of yours 10:21 < bridge> fokkonaut knows more since he made it xd 10:21 < bridge> its about the browser 10:22 < bridge> oh i see 10:22 < bridge> u can show all clients in the server browser 10:22 < bridge> @fokkonaut but ye this is like allocating a std::map or equivalent with 64 slots but it can grow to whathver 10:22 < bridge> https://github.com/ddnet/ddnet/pull/4560/commits/8300753fa41fb90e5042865cf8b2e0fb17a3d292 10:22 < bridge> why use a hash table for a fixed size arr 10:22 < bridge> ``` 10:22 < bridge> pub fn with_capacity(capacity: usize) -> HashSet 10:22 < bridge> 10:23 < bridge> Creates an empty HashSet with at least the specified capacity. 10:23 < bridge> 10:23 < bridge> The hash set will be able to hold at least capacity elements without reallocating. This method is allowed to allocate for more elements than capacity. If capacity is 0, the hash set will not allocate. 10:23 < bridge> ``` 10:23 < bridge> thats nice 10:23 < bridge> Oh at least 10:23 < bridge> https://doc.rust-lang.org/std/collections/struct.HashSet.html#method.with_capacity 10:23 < bridge> mis read 10:23 < bridge> iirc vec grows at a factor of 1.5 10:24 < bridge> ah its 2 10:24 < bridge> so if the initial alloc is 4, to get to 64 it would do 4,8,16,32,64 10:24 < bridge> 5 allocations 10:24 < bridge> vs 1 if u use with capacity 10:25 < bridge> Soon we will get JS client :justatest: 10:25 < bridge> Soon we will get full JS client :justatest: 10:25 < bridge> Hell nah 10:25 < bridge> nice 10:25 < bridge> ddnet ecosystem growing 10:25 < bridge> lots of new libs 10:26 < bridge> :gigachad: 10:26 < bridge> agpl 3 btw :gigachad: 10:27 < bridge> https://teewebs.net 10:27 < bridge> best 10:27 < bridge> teeobserver, ddstats, ddnet-map-diff, ddnet-map-gen are agpl3 too 10:27 < bridge> Its all ur stuff, isnt it? 10:27 < bridge> teerender too 10:27 < bridge> i gotta finish it 10:27 < bridge> ye 10:28 < bridge> @ryozuki u made map diff? 10:28 < bridge> chiller has some kind of map diff tool too, but for github pipeline 10:28 < bridge> @fokkonaut i ported ravie one to rust and added features 10:28 < bridge> its a editor map diff 10:28 < bridge> I have only DDnet-Team-Searcher and web-demo-whateverthehell :feelsbadman: 10:28 < bridge> it modifies the map 10:28 < bridge> And both in js :pepeW: 10:28 < bridge> https://github.com/edg-l/ddnet-map-diff 10:28 < bridge> ye 10:28 < bridge> https://github.com/edg-l/ddnet-map-diff/releases/download/v0.1.1/ddnet-map-diff 10:29 < bridge> linux binary 10:29 < bridge> https://github.com/edg-l/ddnet-map-diff/releases/download/v0.1.1/ddnet-map-diff.exe 10:29 < bridge> and windows 10:29 < bridge> u 10:32 < bridge> keep doing more 10:32 < bridge> i just do whathever idea i have 10:32 < bridge> My ideas take forever 10:33 < bridge> i also have a ddnet config parser 10:33 < bridge> https://github.com/edg-l/teeconfig 10:33 < bridge> ```rust 10:33 < bridge> #[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)] 10:33 < bridge> pub struct ConfigEntry { 10:33 < bridge> /// The description of this config entry. 10:33 < bridge> pub description: String, 10:33 < bridge> /// The type of the entry. 10:33 < bridge> pub entry_type: EntryType, 10:33 < bridge> /// The flags of the entry. 10:33 < bridge> pub flags: CFGFlags, 10:33 < bridge> /// The name of the entry, used in the config file. 10:33 < bridge> pub name: String, 10:33 < bridge> /// The symbol of the entry as used in the source code. 10:33 < bridge> pub symbol: String, 10:33 < bridge> } 10:33 < bridge> ``` 10:33 < bridge> it parses the header c file 10:33 < bridge> and u get a vec with all the config entries with this struct 10:34 < bridge> i need to parse the cfg file next 10:34 < bridge> this is done using a LR(1) parser https://github.com/edg-l/teeconfig/blob/master/src/grammar_cpp.lalrpop 11:05 < bridge> https://frame.work/es/en/products/laptop16-diy-amd-7040 11:07 < bridge> Processor Options 11:07 < bridge> 11:07 < bridge> AMD Ryzenโ„ข 7 7840HS (up to 5.1GHz, 8-core/16-thread, 24MB Cache) 11:07 < bridge> 11:07 < bridge> AMD Ryzenโ„ข 9 7940HS (up to 5.2GHz, 8-core/16-thread, 24MB Cache) 11:07 < bridge> yeah 11:09 < bridge> i wish mmy company gave me one of these 11:09 < bridge> instead of a m1 11:10 < bridge> nerd 11:12 < bridge> all ppl who chat here are nerds 11:12 < bridge> nerds! 11:12 < bridge> https://tenor.com/view/developers-gif-13292051 11:14 < bridge> i really like steve 11:14 < bridge> his uncontested enthusiasm inspires me 11:15 < bridge> tho itโ€™s rly just the cocaine 11:16 < bridge> xd 11:26 < bridge> today i think i get the hardware for my home server 11:26 < bridge> i had a ryzen 3200g lying around so i built it around that 11:26 < bridge> gentoo 2 inc 11:33 < bridge> i not have pc 11:35 < bridge> I know that the package installs curl, so yes 11:35 < bridge> I thought the same a longer time ago, but with twitch support 11:36 < bridge> imagine top ddnet streams coming up in a live feed, or official streams ๐Ÿ™ˆ 11:49 < bridge> curl has nothing to do with websockets 11:49 < bridge> curl is used to query master server and ur name info 11:49 < bridge> http stuff 11:50 < bridge> iirc for websockets we used https://libwebsockets.org/ 12:53 < bridge> > curl has nothing to do with websockets 12:53 < bridge> Calling an api with curl in order to connect the client with the feed would be a valid solution as well. 12:58 < bridge> also I found this, which is also new for me: https://everything.curl.dev/libcurl/ws 12:59 < bridge> also I found this, which is also new for me (can't tell if related): https://everything.curl.dev/libcurl/ws 13:12 < bridge> Pretty powerful for a home server 13:12 < bridge> yeah 13:12 < bridge> I think mine is idle nearly always, except for backups + official ddnet builds + teehistorian archives 13:12 < bridge> but one of the uses ill give it will be to use it with distcc 13:12 < bridge> to compile gentoo packages 13:12 < bridge> xd 13:13 < bridge> @deen do u use some kind of software to make a NAS? 13:13 < bridge> or u just ssh 13:13 < bridge> sshfs 13:13 < bridge> nice 13:14 < bridge> ill check that out 13:14 < bridge> I'm not a Windows user though, probably doesn't work on Windows 13:14 < bridge> yeah im a gentoo user 13:14 < bridge> โ˜• 13:14 < bridge> what other stuff u host in ur home server? 13:14 < bridge> And samba for everyone else in the house 13:14 < bridge> to have some ideas 13:15 < bridge> good thing is i pay a fixed monthly price for water/electricy so i dont care about power usage 13:16 < bridge> Mostly it's a NAS, I also use fetchmail on the server, and get the emails to my local systems using unison. Command line programs that should always run, like newsboat, irssi 13:16 < bridge> ohh 13:16 < bridge> true i can put my irc bouncer there 13:17 < bridge> i also want to do some DIY with my 100 liter aquarium 13:17 < bridge> like a parameter logger and dashboard 13:19 < bridge> <_voxeldoesart> force your fish to use linux 13:19 < bridge> <_voxeldoesart> or smth 13:23 < bridge> ye 13:24 < bridge> ill probs use a microcontroller board to get the params and send them over network to sv 13:24 < bridge> I recommend an rp2040 based one. 13:24 < bridge> https://www.raspberrypi.com/documentation/microcontrollers/rp2040.html 13:25 < bridge> does it has rust tooling/libs 13:25 < bridge> I think so 13:25 < bridge> i have some stm stuff 13:25 < bridge> doing a quick search returns this https://github.com/rp-rs/rp-hal 13:26 < bridge> ah ok ;) 13:28 < bridge> Soon 13:28 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1159089724200329326/screenshot_2023-10-04_13-24-59.png?ex=651e9ded&is=651d4c6d&hm=91f1d28530f603848856cead8ce3df88bd62a94845428391b3aa427d3a4e1b79& 13:30 < bridge> <_voxeldoesart> gotta make that icon 13:30 < bridge> `17 files changed, 985 insertions(+), 435 deletions(-)` :monkaS: 13:31 < bridge> <_voxeldoesart> ooooh i know what type of map robyt3 plays now 13:32 < bridge> <_voxeldoesart> you DO like blocking, dont you, robyt3? 13:32 < bridge> A bit weird how we use the EU "country" for all our European servers but KoG uses individual countries like Germany 13:32 < bridge> Server favorites and friends in my screenshots are always random :kek: 13:33 < bridge> The completed maps are correct though 13:33 < bridge> map* 13:33 < bridge> <_voxeldoesart> ohh 13:33 < bridge> <_voxeldoesart> makes sense 13:35 < bridge> I keep adding random friends for testing. Though when I launch an old client, I lose some of them again, because the maximum number of friends was lower 14:19 < bridge> is that impact font 14:19 < bridge> yes, lol, was the default in paint for some reason 14:19 < bridge> yes, lol, was the default in GIMP for some reason 14:20 < bridge> you could move the set out some scopes so that you don't reallocate in each loop 14:23 < bridge> bitcoin miners? 14:59 < bridge> I'm never touching prediction again :ahendmymiserableexistence: 15:26 < bridge> i tried but the loop complained cuz the references 15:26 < bridge> i can look into it more but i wanted to get smth working fast 15:26 < bridge> could be worth lol 15:26 < bridge> https://nannou.cc/ 15:26 < bridge> @_voxeldoesart look 15:26 < bridge> its Processing in Rust 15:26 < bridge> creative coding art 15:42 < bridge> Will popular unofficial servers be able to get any categorisation? 15:45 < bridge> https://ultraram.tech/ 15:45 < bridge> @chairn opinions on this? 15:45 < bridge> non volatile ram 15:47 < bridge> In theory, yes, but I don't think we discussed the specific criteria adding communities yet 15:47 < bridge> The whole discussion is here: https://github.com/ddnet/ddnet/issues/5654 15:48 < bridge> :justatest: 15:48 < bridge> @robyt3 btw if u ever need a review to get a pr merged feel free to ask 15:51 < bridge> New avatar :brownbear: 15:51 < bridge> who 15:51 < bridge> https://github.com/ddnet/ddnet/pull/7258 15:52 < bridge> btw im in favour of merging this 15:52 < bridge> @heinrich5991 are u against? 15:52 < bridge> curious, it'd have to be something completely different, 100x lower dynamic power consumption than dram usually would imply one tenth the frequency or somehow one one hundreth the capacitance 15:52 < bridge> AIUI, we just merged a PR that fixed a bug with interlaced PNGs? 15:52 < bridge> or there's one in the queue? 15:52 < bridge> what does AIUI mean 15:52 < bridge> as I understand it 15:52 < bridge> <_voxeldoesart> i saw this a year ago 15:52 < bridge> https://github.com/ddnet/ddnet/pull/7283 15:53 < bridge> iwas about to merge it 15:53 < bridge> but then i read it was useful only with that pr 15:53 < bridge> > Note that the client will still show a warning popup when loading images with an indexed color palette, so this is only effectively usable together with #7258. 15:53 < bridge> https://github.com/ddnet/ddnet/pull/7258 15:53 < bridge> <_voxeldoesart> tried ninnou but then again its as as hell 15:53 < bridge> it should be useful without the PR as well 15:53 < bridge> no idea why robyte said that 15:53 < bridge> @robyt3? 15:54 < bridge> I guess it's still an improvement to fix the loading of the images itself, but we wouldn't add our own images if they still cause the warning every time 15:55 < bridge> so https://github.com/ddnet/ddnet/pull/7283 good to merge because it fixes a bug, I'd say 15:55 < bridge> well ill add it to queue 15:55 < bridge> yeah 15:57 < bridge> https://github.com/ddnet/ddnet/pull/6990 15:57 < bridge> lgtm 15:58 < bridge> I think I'd like to wait with https://github.com/ddnet/ddnet/pull/7258, people are still actively using clients that don't support these PNGs 15:58 < bridge> I think I'd like 7258 merged, one more mistake like this and they will just have our heads 15:58 < bridge> we shouldnt hold back because 20 people use ancient clients 15:59 < bridge> if you talk to these people and understand why they use these clients, we can figure something out 15:59 < bridge> we probably also "hold back" because 20 people use some feature in other places 15:59 < bridge> e.g. adding upnp support to the servers 15:59 < bridge> or re-adding freeze stars 15:59 < bridge> There will be still a couple people that just prefer to stay on the old client, even if we convert most of them you won't agree to merging 7258 because of the holdouts 15:59 < bridge> usually these ppl hold back not because they use a feature but because smth changed 15:59 < bridge> even konsti uses latest client 15:59 < bridge> he told me xd 16:00 < bridge> I find statistics liek "only 1% of our concurrent player base use the feature" to be misleading 16:00 < bridge> i think ppl using old clients understand what it means 16:00 < bridge> and the drawbacks is that at some point that old version wont work with modern server versions 16:01 < bridge> but we're cutting an online service which means we stop supporting the skin db for these skins in these old clients 16:01 < bridge> I think you should provide a patch to make the warning far less punishing if you don't want 7258. We don't get to kill peoples runs for our mistake 16:01 < bridge> I guess I'll try to make a patch for that 16:01 < bridge> I can see this point 16:02 < bridge> (that is at the very least, I think 7258 + a skindb policy just to never include any interlaced or indexed skin is the better solution) 16:02 < bridge> if you believe we can do a skindb policy without hte warnings, then the warnings shouldn't be a problem 16:03 < bridge> I don't believe in that, that's why I added the warning 16:03 < bridge> we can *uphold the skindb policy 16:04 < bridge> I'm starting to wonder why you ever added the disable switch to the warning if you'd never be willing to disable it anyway 16:04 < bridge> I think it's not the time yet 16:05 < bridge> I added it because otherwise the warning would be a forward compatibility hazard 16:05 < bridge> What is the bar if not 1%? 0.1%? 20 players? 10 players? 16:06 < bridge> Why? If we are never allowed to break backwards compatibility, in this case there will never be a forward compatibility issue anyway 16:07 < bridge> @deen any opinion 16:07 < bridge> if we didn't add the option, we'd be in the same situation as before 16:07 < bridge> Anyway, as long as the warning is toned down, I'm okay with keeping it if you feel it's that important. My concern is mostly with punishing the most players, the rest is just idle curiosity as to what you are thinking 16:07 < bridge> at least now we're already in a much better position than before, because we added the remote switch 16:08 < bridge> Anyway, as long as the warning is toned down, I'm okay with keeping it if you feel it's that important. My concern is mostly with punishing most players, the rest is just idle curiosity as to what you are thinking 16:10 < bridge> Well the switch might as well not exist if we are not allowed to touch it. The mere existence of it only gives us an option, if we aren't allowed to take the option it's neutral 16:10 < bridge> > the rest is just idle curiosity as to what you are thinking 16:10 < bridge> I think stopping software that once worked from working is bad, especially if people are still using it 16:11 < bridge> I agree that stopping it is bad. One skin not properly loading in 3 years of hosting the db, almost 5 years since we released the said old version does not constitute stopping it IMO 16:11 < bridge> I agree that this is kinda minor, it's just skins 16:12 < bridge> If it was a crash or graphical issues I'd be 100% onboard 16:12 < bridge> It falls back gracefully to the default skin even 16:12 < bridge> yes 16:13 < bridge> one skin is not bad, I agree 16:13 < bridge> I'd guess that it's just one skin because we had the policy in place, perhaps even due to the warning 16:14 < bridge> I guess my fear is that by removing the warning, we're going to get to a point where the skindb is not usable for old clients 16:14 < bridge> still used old clients 16:14 < bridge> so just removing the warning in the current state isn't bad 16:15 < bridge> Is there even a will to include interlaced or indexed skins in the db? 16:15 < bridge> I saw such a will from people, almost just to exclude old clients ^^ 16:15 < bridge> it'd also give some space savings 16:16 < bridge> Oh that is definitely there, @ryozuki and @jupeyy_keks would interlace all of them just to make people migrate ๐Ÿ˜„ 16:18 < bridge> This is so interesting btw, it's like a space-age FLOTOX ๐Ÿ˜„ 16:19 < bridge> It uses gallium arsenide semiconductors too, bizarre 16:19 < bridge> I don't care much, probably because I think seeing a default skin is not that ad. Like Learath2 said let's set a reasonable user base where we accept a few skins breaking, maybe 1%. 16:22 < bridge> xD 16:23 < bridge> https://github.com/WebAssembly/stringref/blob/a64917cd5346f8704e614c4825ebf05737ac5e64/proposals/stringref/Overview.md 16:23 < bridge> oof, encoding broken utf-16 into another standard ๐Ÿ˜ฆ 16:25 < bridge> How loose the web technologies are with standards is actually quite annoying. Browsers are so extremely liberal with what they accept that it's sometimes hard to predict behaviour 16:25 < bridge> causing security issues in the process 16:25 < bridge> Potentially ill-formed UTF-16 should just look broken so the developer fixes it, not patched up and displayed 16:28 < bridge> the largest languages all use this potentially ill-formed utf-16, unfortunately ๐Ÿ˜ฆ 16:34 < bridge> but maybe this gives you better string interaction for building wasm APIs ๐Ÿ™‚ @ryozuki 16:42 < bridge> <_voxeldoesart> ironically the biggest ddnet update will be adding backwards compat 16:43 < bridge> :poggers2: 16:43 < bridge> @learath2 i think js Date also uses iso86xx with quirks 16:43 < bridge> i forgot the full iso name 16:54 < bridge> > https://maggiepint.com/2017/04/11/fixing-javascript-date-web-compatibility-and-reality/ 17:16 < bridge> is there support for gamemodes on the ddnet codebase? or is the DDRace gamemode hardcoded? I'm trying to make a new gamemode and looking at other mods (which aren't based on ddnet codebase) seem to do it differently than DDRace. 17:20 < bridge> Many things are hard-coded in our code. There's another project that tries to make the code more modable: https://github.com/teemods/ddnet-skeleton 17:22 < bridge> Ok, thanks ^^ 17:26 < bridge> <_voxeldoesart> oh man you know what would be a cool af mod support feature? the ability to have popups that have buttons and prompts that can send what button you press or what script you type in to the server 17:51 < bridge> Dead/not maintained codebase which starts from disabling tests. Also it is hard to agree what is the base game (and should be kept) and what is ddnet specific stuff (and should be cut). 17:51 < bridge> It could be a thing if we'd have more people (mod devs) working on the shared codebase. I tried to convince a few mod devs but they see it as waste of time/resources. 18:00 < bridge> Dead/not maintained codebase which starts with disabling tests. Also it is hard to agree what is the base game (and should be kept) and what is ddnet specific stuff (and should be cut). 18:00 < bridge> It could be a thing if we'd have more people (mod devs) working on the shared codebase. I tried to convince a few mod devs but they see it as waste of time/resources. 18:05 < bridge> IIUC this is the goal of https://github.com/ddnet/ddnet/issues/7042 (which uses https://nodes.teeworlds.dev/media as an example) 18:05 < bridge> However, the proposed "strategy" as "implement an interface in C++ and subsequently add Lua bindings for it" is questionable, and there are other ways to do that. 18:44 < bridge> @nuborn if you have some time, could you check out reckoning in tutorial? It's broken because m_Jumps is not correct. I think we are forced to take that though now that I see we are kinda stuck with at most what vanilla can reckon 18:44 < bridge> yes, I could take a look later 18:58 < bridge> @learath2 do u know if u can make the gentoo install boot and setup ssh? without any display/keyboard 18:58 < bridge> :justatest: 18:58 < bridge> ill have to connect it to a display and keyboard right? smh 18:59 < bridge> quite pretty and smol i would say 18:59 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1159173057458491463/IMG_20231004_185827.jpg?ex=651eeb89&is=651d9a09&hm=16d79c389b99abc39da6e7737061f8affce7024a1f7691d0bc1485672b302609& 19:06 < bridge> it's doable if you have dhcp and using an ethernet cable 19:08 < bridge> @learath2 u mean the router assigning the computer a local ip? 19:08 < bridge> and yes i got ethernet 19:11 < bridge> yes, in that case the installation cd can bring up the network on it's own 19:11 < bridge> You can just symlink sshd into `/etc/runlevels/boot` in the install usb and it should bring up sshd 19:18 < bridge> (though doing it blindly can be a fun exercise, it can also be extremely frustrating as you have no way to monitor anything ๐Ÿ˜„ ) 19:20 < bridge> lmao 19:44 < bridge> https://ferrous-systems.com/blog/ferrocene-open-source/ 19:44 < bridge> > Ferrocene is the main Rust compiler - rustc - but quality managed and qualified for use in automotive and industrial environments (currently by ISO 26262 and IEC 61508) by Ferrous Systems. It operates as a downstream to the Rust project, further increasing its testing and quality on specific platforms. We will publish further detailed blog posts on the components and changes in Ferrocene in the coming weeks. 19:51 < bridge> gods 20:00 < bridge> ``` 20:00 < bridge> โฏ sudo dd if=install-amd64-minimal-20230924T163139Z.iso of=/dev/sdc bs=8192k; sync 20:00 < bridge> 58+1 records in 20:00 < bridge> 58+1 records out 20:00 < bridge> 488900608 bytes (489 MB, 466 MiB) copied, 1.43392 s, 341 MB/s 20:00 < bridge> ``` 20:00 < bridge> holy shit that was fast 20:00 < bridge> good shit my usb is fast 20:00 < bridge> it did it almost instant 20:01 < bridge> Oh with sync even, I was about to say there is a chance you wrote it all to a ram buffer ๐Ÿ˜„ 20:01 < bridge> xd 20:01 < bridge> one day i was sick of slow usbs 20:02 < bridge> so i bought the fastest usb i found 20:02 < bridge> also its 128gb 20:02 < bridge> sandisk extreme pro 20:04 < bridge> how to do this btw 20:04 < bridge> i guess i need to mount the usb 20:04 < bridge> yep 20:05 < bridge> OH also remove pwgen from whatever runlevel it's in 20:05 < bridge> and set a root password instead 20:07 < bridge> how was it to mount with write perms 20:08 < bridge> `mount -o rw` ? 20:08 < bridge> source write-protected, mounted read-only. 20:08 < bridge> ``` 20:08 < bridge> gentoo /mnt # mount -o rw UUID=2023-09-24-18-32-47-00 /mnt/liveusb/ 20:08 < bridge> mount: /mnt/liveusb: WARNING: source write-protected, mounted read-only. 20:08 < bridge> gentoo /mnt # mount -o remount,rw UUID=2023-09-24-18-32-47-00 /mnt/liveusb/ 20:09 < bridge> mount: /mnt/liveusb: cannot remount /dev/sdc read-write, is write-protected. 20:09 < bridge> dmesg(1) may have more information after failed mount system call. 20:09 < bridge> ``` 20:09 < bridge> oh 20:09 < bridge> im trying to mount the device xd 20:09 < bridge> sec 20:10 < bridge> ``` 20:10 < bridge> sdc 8:32 1 119.3G 0 disk /mnt/liveusb iso9660 ISOIMAGE 2023-09-24-18-32-47-00 20:10 < bridge> โ”œโ”€sdc1 8:33 1 246K 0 part 99354d5d-f71e-4501-a85f-556db6b5e0c5 20:10 < bridge> โ”œโ”€sdc2 8:34 1 2.8M 0 part vfat 6B18-43A4 99354d5d-f71e-4501-a85c-556db6b5e0c5 20:10 < bridge> โ”œโ”€sdc3 8:35 1 462.9M 0 part hfsplus ISOIMAGE 99354d5d-f71e-4501-a85d-556db6b5e0c5 20:10 < bridge> โ””โ”€sdc4 8:36 1 300K 0 part 99354d5d-f71e-4501-a85a-556db6b5e0c5 20:10 < bridge> ``` 20:10 < bridge> xd 20:10 < bridge> can u mount by partuuid 20:10 < bridge> why not just `/dev/sdc3`? 20:11 < bridge> gentoo /mnt # mount /dev/sdc3 liveusb/ 20:11 < bridge> mount: /mnt/liveusb: unknown filesystem type 'hfsplus'. 20:11 < bridge> dmesg(1) may have more information after failed mount system call. 20:11 < bridge> hmm 20:11 < bridge> xd 20:11 < bridge> Yeah I guess this bizarre partition type is just not easy to modify 20:11 < bridge> sudo mount -t hfsplus -o rw,remount -force /dev/sdb3 20:11 < bridge> i found this command 20:11 < bridge> tough luck, you'll have to just connect a monitor and keyboard ๐Ÿ˜„ 20:12 < bridge> f 20:12 < bridge> Why hfsplus though, I don't really believe that 20:12 < bridge> I think it's a bizarre result of this hybrid iso disk image thing 20:13 < bridge> ye 20:14 < bridge> sad noises 20:14 < bridge> I'm curious, could you run `cfdisk` on the iso? 20:14 < bridge> holy 20:15 < bridge> whats cfdisk 20:15 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1159191976944611374/image.png?ex=651efd28&is=651daba8&hm=597ae7c40befca8692610e6725f15af532f33d99af2bf9091f5977f75336ae15& 20:15 < bridge> it's fdisk but better 20:15 < bridge> why HFS+? that's so bizarre 20:15 < bridge> xd 20:16 < bridge> Ah, maybe for mac compatibility? 20:16 < bridge> oh probs 20:36 < bridge> <_voxeldoesart> take a shot every time someone uses "blazingly" as a buzzword when advertizing rust 20:36 < bridge> <_voxeldoesart> https://cdn.discordapp.com/attachments/293493549758939136/1159197287604572202/Screenshot_20231004_143518_Chrome.jpg?ex=651f021a&is=651db09a&hm=13fa6073ab88d5f585333c78c6ddc74779dd2022504757bb63e83ad87998a68c& 20:41 < bridge> alcohol coma 20:41 < bridge> @learath2 should i use btrfs 20:41 < bridge> btw cfdisk is pog 20:42 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1159198842865057892/rn_image_picker_lib_temp_e1244dfc-4e9b-4d87-bc61-ef22e7695253.jpg?ex=651f038d&is=651db20d&hm=352ef7aa5f5c8e104d5a8b89be0d6fd344de6b3a6a6cb1ac1e9e17d887600ddf& 20:42 < bridge> I used it for the lulz a couple times, but never for anything serious, I vaguely remember @heinrich5991 using it for more serious things 20:42 < bridge> I still just go for ext4 20:42 < bridge> i guess same 20:43 < bridge> I tried zfs for a NAS once and it is quite nice 20:43 < bridge> im proud i did the filesystem without looking at the guide 20:43 < bridge> i already remember all cmds xd 20:44 < bridge> What are you doing btw? New computer? 20:44 < bridge> home server 20:44 < bridge> with gentoo 20:44 < bridge> ill help it with distcc 20:44 < bridge> altho it has a 3200g ryzen 20:44 < bridge> 8 cores 20:44 < bridge> iirc 20:44 < bridge> everything needs to be gentoo 20:44 < bridge> i cant dare touch another distro in this home 20:45 < bridge> :gigachad: 20:45 < bridge> Try a BSD sometime, it's quite interesting too 20:46 < bridge> Portage is inspired by ports iirc 20:48 < bridge> should i use no multilib 20:48 < bridge> for a server 20:48 < bridge> iirc i only use multilib on desktop cuz steam 20:49 < bridge> Mh, why make it annoying for yourself? Portage won't pull in anything 32bit anyway unless you want it 20:52 < bridge> <_voxeldoesart> what does ryoz need a server for 20:52 < bridge> ye i went normal 20:53 < bridge> its always fun and its handy to have a computer running 24/7 20:53 < bridge> Having a homeserver is quite fun 20:53 < bridge> I used to have one hosting all my anime that I totally acquired through legitimate means 20:53 < bridge> <_voxeldoesart> i mean sure but like 20:53 < bridge> <_voxeldoesart> what plans would you have 20:53 < bridge> NAS, irc bouncer 20:54 < bridge> torrent seeder (?) 20:54 < bridge> more to come 20:54 < bridge> it can also help my main pc compule 20:54 < bridge> compile 20:54 < bridge> via distcc 20:54 < bridge> im lookijg forward to halved firefox times 20:54 < bridge> or almosy 20:54 < bridge> almost* 20:54 < bridge> why sweat emoji? 20:55 < bridge> someone doesnโ€™t understand torrenting :justatest: 20:55 < bridge> you can also have your server compile firefox for your main machine 20:55 < bridge> just in the background 20:55 < bridge> omg 20:55 < bridge> thats genius 20:55 < bridge> or everything for a fact 20:55 < bridge> xsd 20:55 < bridge> xdd 20:55 < bridge> <_voxeldoesart> too used to "getting games free" 20:59 < bridge> You can have your server just compile everything make it into a binpkg and use your server as a binpkg host on your local machine ๐Ÿ˜„ 20:59 < bridge> That's how I was using gentoo on my laptop back in the day 21:51 < bridge> afair torrent speeds are always higher than average fileshare service's 22:08 < bridge> Especially if you have an above average connection 22:09 < bridge> wonder if i can run distcc on winders 22:11 < bridge> you probably wouldn't want to do this on your home internet, there are seedbox-for-rent VPS services that are just underpowered with a lot of storage and a lot of bandwidth 22:11 < bridge> the bandwidth is what makes it worth it 22:12 < bridge> they can get up to like 50Gbps 22:12 < bridge> but are usually around 10-20 22:13 < bridge> like if i hosted a seedbox at my house, my ISP's disgustingly asymmetrical grid would make it so not worth it 22:14 < bridge> less than 5 MBps 22:22 < bridge> it looks too beautiful to be true. Feels more like a scam to get more money from investors. 22:22 < bridge> > Fabricated single-cell memories show clear 0/1 logic-state contrast after โ‰ค10 ms duration program/erase pulses of โ‰ˆ2.5 V 22:22 < bridge> Good luck to sell that to anyone 22:22 < bridge> Everyone in the semiconductor expects a non volatile memory with the DRAM performance. None has yet achieved that, mostly because of endurance 22:24 < bridge> > 10 ฮผm gate-length device 22:24 < bridge> great, just 1000x bigger than competitors 22:25 < bridge> > Conflict of Interest 22:25 < bridge> > The authors declare no conflict of interest 22:25 < bridge> yeah sure 22:27 < bridge> Baldurs Gate 3 has the exact same screenshot issue we have in ddnet ๐Ÿ˜„ 22:27 < bridge> > ultralow logic state switching energies of 10โˆ’17 J are predicted for 20 nm feature size ULTRARAM memories 22:27 < bridge> Predicted, they admit they don't have the tech ready yet, which confirms my first point. They need money from investors 22:30 < bridge> Hm, it seems AMD uses something atleast named the same in some devices https://docs.xilinx.com/r/en-US/am007-versal-memory/UltraRAM-Introduction 22:32 < bridge> looks completely unrelated. Your link seems to refer to FPGA only. It doesn't even mention it's non volatile 22:33 < bridge> Their concept is closer to Intel Optane memories 22:34 < bridge> I was following a link from wikipedia, so I guess whoever the editor is just googled ultraram and added a source ๐Ÿ˜„ 22:34 < bridge> 22:34 < bridge> Ultraram SRAM block array, not even DRAM 22:36 < bridge> Another half scam company https://www.weebit-nano.com/ 22:36 < bridge> It does work, but only for niche market with specific usage, not for everyday desktop 22:36 < bridge> Q: Won't this be the case until someone does achieve it? I don't quite see this as an argument for much of anything 22:37 < bridge> Unless you believe there is some fundamental physics reason that makes it actually impossible 22:37 < bridge> If it was true, everyone here would hear about it 22:37 < bridge> not really, but the techs used have troubles to get good performance, endurance and density 22:37 < bridge> in general, you only get 2 out of 3 22:38 < bridge> SRAM and DRAM have good perf and endurance, but quite low density, and also quite high power consumption 22:38 < bridge> Flash has bad performance but very high density. High power consumption in writing, low in reading. Endurance is quite limited 22:39 < bridge> New non volatile memories often have good performance, high density, low/medium power consumption, but low endurance as well 23:09 < bridge> @robyt3 do you have any idea how to refactor the layer list dragging code? I tried multiple times but it just doesnt work. Ideally it would be reusable for stuff like the server settings editor (or envelopes :justatest:). So the part you drag isnt always the same (ie in the layer list its the group button, for settings it would be a `โ ฟ` label), which doesn't make it easier. 23:21 < bridge> I don't have any immediate ideas, but I also think it would be useful to refactor the dragging UI code. For example the usage of `DoDraggableButtonLogic` feels a bit inconvenient with the two bool pointer arguments. Dragging/moving support could be added to listboxes in general and enabled when desired if possible. I think making the entire list item draggable would be easier for server settings. I supposed you could also add delete buttons to each 23:26 < bridge> Allowing the whole line do be draggable seems good too. Adding support it to listbox would be even better. One of the biggest problems I faced was the hot/active system. It doesnt really allow you to have subelements. So say I wanted to add a movable container. Then it would be hot if the mouse is hovering it. But that completely breaks all buttons that are potentially in the container. 23:26 < bridge> Allowing the whole line do be draggable seems good too. Adding support to listbox would be even better. One of the biggest problems I faced was the hot/active system. It doesnt really allow you to have subelements. So say I wanted to add a movable container. Then it would be hot if the mouse is hovering it. But that completely breaks all buttons that are potentially in the container. 23:37 < bridge> gg i got my home server booted and sshed